-
Novel Possibilities for Advanced Molecular Structure Design
of Polymers and Networks
Anna Finne
Department of Fibre and Polymer Technology Royal Institute of
Technology
Stockholm, Sweden 2003
Akademisk avhandling
som med tillstånd av Kungliga Tekniska Högskolan framlägges för
offentlig granskning för avläggande av teknisk doktorsexamen
fredagen den 31 oktober 2003, kl 9.00 i V1, Teknikringen 76, KTH,
Stockholm. Avhandlingen försvaras på engelska.
-
ISBN 91-7283-577-X
Printed by Universitetsservice US AB, Stockholm, Sweden
-
.
Om du försöker hitta Hem men istället hittar en
grop, kan du försöka att leta efter en grop. Då
skulle du med all säkerhet inte hitta en grop, vilket skulle
vara bra, för då
kanske du hittar någonting du inte letar efter, vilket skulle
kunna vara precis
det du letar efter
-
Novel Possibilities for Advanced Molecular Structure Design of
Polymers and Networks
Anna Finne Department of Fibre and Polymer Technology, Royal
Institute of Technology,
Stockholm, Sweden
ABSTRACT Synthetic and degradable polymers are an attractive
choice in many areas, since it is possible to control the way in
which they are manufactured; more specifically, pathways to
manipulate the architecture, the mechanical properties and the
degradation times have been identified. In this work, L-lactide,
1,5-dioxepan-2-one and ε-caprolactone were used as monomers to
synthesize polymers with different architectures by ring-opening
polymerization. By using novel initiators, triblock copolymers,
functionalized linear macromonomers and star-shaped aliphatic
polyesters with well-defined structures have been synthesized. To
synthesize triblock copolymers, cyclic germanium initiators were
studied. The polymerization proceeded in a controlled manner
although the reaction rates were low. To introduce functionality
into the polymer backbone, functionalized cyclic tin alkoxides were
prepared and used as initiators. During the insertion-coordination
polymerization, the initiator fragment consisting mainly of a
double bond was incorporated into the polymer backbone. The double
bond was also successfully epoxidized and this gave unique
possibilities of synthesizing graft polymers with precise spacing.
The macromonomer technique is a very effective method for producing
well-defined graft polymers. Spirocyclic tin initiators were
synthesized and used to construct star-shaped polymers. The
star-shaped polymers were subsequently crosslinked in a
polycondensation reaction. These crosslinked structures swelled in
water, and swelling tests showed that by changing the structure of
the hydrogel network, the degree of swelling can be altered. A
first evaluation of the surface characteristics of the linear
triblock copolymers was also performed. AFM analysis of the
heat-treated surfaces revealed nanometer-scale fibers and tests
showed that keratinocytes were able to grow and proliferate on
these surfaces.
Keywords: ring-opening polymerization, coordination-insertion,
germanium, cyclic tin alkoxides, spirocyclic initiators,
poly(L-lactide), poly(1,5-dioxepan-2-one), triblock, star-shaped,
network, functionalization, morphology, AFM
-
SVENSK SAMMANFATTNING De vetenskapliga framsteg inom cellbiologi
och syntetiska polymermaterial de senaste åren har öppnat nya
möjligheter för avancerad implantat kirurgi. Eftersom behovet av
implantat idag är så stort att det inte kan tillgodoses av
mänskliga donatorer är det viktigt att sjuka och skadade organ
ersätts av artificiell vävnad. En mycket lovande teknik för
vävnadsrekonstruktion är tissue engineering. Det är en metod som
innebär att vävnadsspecifika celler får växa, in vitro eller in
vivo, över ett polymert bärarmaterial. Polymeren upptar under en
uppbyggnadsfas den mekaniska belastningen, men gradvis resorberas
bärarmaterialet då kroppens egen vävnad återbildas. För att detta
skall fungera måste det polymera materialets egenskaper vara mycket
specifika och kontrollerbara. En stor del av detta projekt har
syftat till att syntetisera resorberbara polymera material med
specifik och funktionell arkitektur med hjälp av
ringöppningspolymerisation och med nya initiatorer som verktyg. Det
skall skapa möjligheter att framställa syntetiska analoger till
naturens hierarkiska biomaterial; nedbrytbara polymerer med
väldefinierad struktur och egenskaper skräddarsydda för att
uppfylla alla de krav som ställs på syntetiska implantat i
verkliga, biomedicinska applikationer.
Triblock sampolymerer, makromonomerer funktionaliserade med en
dubbelbindning samt stjärnformade polymerer har syntetiserats
kontrollerat. Dessa har i sin tur använts i efterföljande
reaktioner för att skapa mer avancerade strukturer.
− Att använda funktionaliserade makromonomerer där
reaktionspunkterna är förutbestämda för att syntetisera nya
strukturer är mycket användbart. För att utröna dubbelbindningens
reaktivitet i förgreningsreaktioner utfördes epoxideringstest. Utan
att påverka resterande delar av huvudkedjan bildades epoxider,
vilka kan användas för att skapa förgrenade polymerer eller
nätverk.
− För att undvika problem vid implantation bör implantatet likna
kroppens egen vävnad så mycket som möjligt. Det är viktigt att allt
från mekaniska egenskaper till morfologi är optimerat och passar
den aktuella applikationen. I detta projekt har hydrogeler
syntetiserats utifrån de stjärnformade polymererna. En hydrogel som
är hydrofil sväller i vatten och blir mycket mjuk och formbar utan
några vassa kanter, vilka skulle kunna orsaka irritation.
Svällningsegenskaperna kunde påverkas genom att använda olika
monomersammansättningar och tvärbindningsgrad i polymeren.
− En god interaktion mellan substrat och levande celler är
fundamentalt inom tissue engineering. Andra forskargrupper har
visat att substratets topografi och morfologi påverkar cellernas
utbredning påtagligt och att det går att styra cellernas
tillväxtriktning med hjälp av olika mönster i substratet. För att
finna vägar till att stimulera cellers adhesion, spridning och
orientering har materialens ytor studerats
-
och försök att påverka ytan har genomförts. Vi har funnit att
det bildas unika fiberstrukturer då triblockpolymererna
fasseparerar vid värmebehandling. Detta arbete visade att fibrerna
ligger parallellt och att det går att påverka bland annat
tjockleken på dessa fibrer. Ytorna var inte cytotoxiska.
Celltillväxten på ytorna var god och varierade i form och
utsträckning beroende på polymerernas sammansättning.
-
LIST OF PAPERS This thesis is a summary of the following
papers:
I. "Use of Germanium Initiators in Ring-Opening Polymerization
of L-Lactide"
A. Finne, Reema and A. C. Albertsson J. Polym. Sci., Part A:
Polym. Chem. 2003, 41, 3074-3082
II. "L-Lactide Macromonomer Synthesis Initiated by New Cyclic
Tin Alkoxides Functionalized for Brushlike Structures" M. Ryner, A.
Finne, A. C. Albertsson and H. R. Kricheldorf Macromolecules 2001,
34, 7281-7287
III. "New Functionalized Polyesters to Achieve Controlled
Architectures" A. Finne and A. C. Albertsson J. Polym. Sci., Part
A: Polym. Chem. accepted for publication
IV. "Controlled Synthesis of Star-Shaped L-Lactide Polymers
Using New
Spirocyclic Tin Initiators" A. Finne and A. C. Albertsson
Biomacromolecules 2002, 3, 684-690
V. "Polyester Hydrogels with Swelling Properties Controlled by
the Polymer
Architecture, Molecular weight, and Crosslinking Agent" A. Finne
and A. C. Albertsson J. Polym. Sci., Part A: Polym. Chem. 2003, 41,
1296-1305
VI. "Well-Organized Phase-Separated Nanostructured Surfaces
of
Hydrophilic/Hydrophobic ABA Triblock Copolymers" A. Finne, N.
Andronova, A. C. Albertsson Biomacromolecules 2003, 4,
1451-1456
-
TABLE OF CONTENTS 1. PURPOSE OF THE STUDY
............................................. 1
2. INTRODUCTION
.......................................................... 3
2.1 Background
.....................................................................................
3 2.2
Monomers........................................................................................
4
2.2.1 L-Lactide (LLA)
..............................................................................................4
2.2.2 ε-Caprolactone (ε-CL)
....................................................................................5
2.2.3 1,5-Dioxepan-2-one
(DXO)............................................................................6
2.3 Ring-opening polymerization
..................................................... 6 2.3.1
Coordination-insertion mechanism
.............................................................7
2.3.2 Ring-opening polycondensation
..................................................................8
2.4 Macromolecular
design.................................................................
8 2.4.1
Copolymerization...........................................................................................8
2.4.2 Functionalized
macromonomers..................................................................9
2.4.3 Star-shaped polymers
..................................................................................12
2.5
Networks........................................................................................
12 2.5.1
Hydrogels......................................................................................................13
2.6 Biomedically adapted
surfaces.................................................. 14 2.6.1
Cell adhesion – hydrophilicity and hydrophobicity
...............................14 2.6.2 Cell adhesion - morphology
and topography ..........................................15
3.
EXPERIMENTAL.........................................................
17
3.1 Materials
........................................................................................
17 3.2 Synthesis of initiators
.................................................................
17
3.2.1 Germanium initiators
..................................................................................17
3.2.2 Functionalized tin
initiators........................................................................18
3.2.3 Spirocyclic tin
initiators...............................................................................18
3.3 Polymerization model reaction
................................................. 18 3.4
Epoxidation
...................................................................................
19 3.5
Copolymerization.........................................................................
19
-
3.6 Synthesis of networks
.................................................................
19 3.6.1 Tetra-functional acid chloride
....................................................................19
3.6.2 Crosslinking
reaction...................................................................................19
3.7 Film
preparation...........................................................................
20 3.8 Characterization
methods...........................................................
20
3.8.1 Nuclear Magnetic Resonance
.....................................................................20
3.8.2 Size Exclusion Characterization
.................................................................20
3.8.3 Differential Scanning
Calorimetry.............................................................21
3.8.4 Atomic Force Microscopy
...........................................................................21
3.8.5. Environmental Scanning Electron
Microscopy.......................................21 3.8.6 Swelling
.........................................................................................................21
3.9 Cell response measurements
..................................................... 22 3.9.1
Time-lapse
videomicroscopy......................................................................22
4.TRIBLOCK
COPOLYMERS.............................................. 23
4.1 Germanium
initiators..................................................................
23 4.2
Polymerization..............................................................................
27 4.3 Thermal characteristics
...............................................................
30
5. FUNCTIONALIZED POLYESTERS ..................................
33
5.1 Functionalized cyclic tin(IV)alkoxides
.................................... 33 5.2 Synthesis of
functionalized polyesters.................................... 35
5.3 Epoxidation of the incorporated doublebond
........................ 39
6. STAR-SHAPED
POLYESTERS........................................ 45
6.1 Spirocyclic tin
alkoxides.............................................................
45 6.2 Synthesis of star-shaped poly(L-lactide)
................................. 46
7. NETWORKS
...............................................................
51
7.1 Synthesis of networks
.................................................................
51 7.2 Characterization of
networks.....................................................
53
8. SURFACE CHARACTERIZATION ...................................
57
8.1 Influence of process
parameters................................................ 58
-
8.1.2 Variation of the segment lengths
...............................................................60
8.2 Topography
...................................................................................
61 8.3 Cell adhesion
................................................................................
62
9. CONCLUSIONS
.......................................................... 67
10. FUTURE PERSPECTIVES
............................................ 69
11. ACKNOWLEDGEMENTS
............................................ 71
12. REFERENCES
........................................................... 73
-
ABBREVIATIONS AFM atomic force microscopy DS degree of swelling
[%] DSC differential scanning calorimetry DXO 1,5-dioxepan-2-one
ε-CL ε-caprolactone ECM extracellular matrix ESEM environmental
scanning electron microscope HOMO highest occupied molecular
orbital LLA L-lactide LUMO lowest unoccupied molecular orbital MWD
molecular weight dispersity NMR nuclear magnetic resonance PLLA
poly(L-lactide) PEG poly(ethylene glycol) ROP ring-opening
polymerization SEC size exclusion chromatography SEM scanning
electron microscope mCPBA m-chloroperoxybenzoic acid
SYMBOLS DP degree of polymerization [I] initiator concentration
[M] [M] monomer concentration [M] [M]0 initial monomer
concentration [M] Mn number-average molecular weight [g/mol] Tc
crystallization temperature [°C] Tg glass transition temperature
[°C] Tm melting temperature [°C] W final weight [g] Wo initial
weight [g]
-
Purpose of the study
1
1. PURPOSE OF THE STUDY The present work discusses the research
reported in the appended publications. The work was designed to
provide a broader experimental basis for a final conclusion as to
whether or not new initiators can be used to synthesize
well-defined usable polymers with different architectures.
Synthetic polymers have found an enormous amount of different
applications during recent decades. Every application requires
specific properties. One major problem is that the synthetic
polymers are relatively simple compared to the polymers built by
nature. An objective of molecular architecture, and also of this
work, is to design new specialized polymers in a controlled way.
Studies have been made to assess the use of new initiators for
obtaining functionalized polymers as well as polymers with advanced
and controlled architectures. Three different types of initiators
have been used:
− Germanium initiators
− Functionalized cyclic tin alkoxide initiators
− Spirocyclic tin alkoxide initiators
Subsequent reactions of the synthesized macromonomers and
polymers were studied in order to determine their usefulness.
Epoxidation and crosslinking reactions were carried out using the
functionalized macromonomers and the star-shaped polymers
respectively.
The potential application of polymers synthesized in this work
is in the biomedical field. The morphology and topography of the
material are therefore decisive since cell adhesion and spreading
are influenced by the physico-chemical characteristics of the
underlying substrate. Some surface characterization and cell growth
studies have been performed as a part of this work.
-
Introduction
3
2. INTRODUCTION 2.1 Background
Synthetic polymers are nowadays used all around us. Their
applications range from electrical conductors to scaffolds in
tissue engineering. Properties such as hydrophilicity, degradation
rate and mechanical properties have to be optimized in relation to
the envisioned application. There are various ways to influence
these properties; the most common being by copolymerization or by
synthesizing functionalized polymers with specific architectures.
The ability to control and reproduce the reactions is extremely
important in every method. Different characterization methods make
it possible to obtain the exact structure of the material and to
relate the material properties to this structure. Thus it is
possible to design the synthetic approach for achieving the best
material properties for the desired application. In the medical
field, the need for specialized materials with controlled
properties is high. Loss or failure of organs as a result of an
injury or other type of damage is a growing human health problem.
The health care costs in North America for tissue loss or end-stage
organ failure exceed $400 billion per year.1 There are not enough
donors and the need for synthetic alternatives is growing. Tissues
and organs consist of living cells arranged within a framework
called the extracellular matrix (ECM). The ECM is a gel composed of
proteins and polysaccharides, and it plays an important role during
growth and wound repair. It meshes the cells together and serves as
a reservoir for the signaling molecules that control the migrating,
proliferating, and differentiating cells. It also provides
strength, rigidity, cellular communication, cellular protection and
transport of nutrients and hormones. The ECM acts as the
communication highway between cells and other extracellular fluids.
Because of the local differences in the composition and
organization of the ECM tendons resist tension and cartilage resist
compression. Degeneration of ECM affects, for example, the ability
to effectively assimilate nutrients into the cell, it will result
in a slowed repair of damaged tissue with increased scar tissue
deposition. Artificial substitutes for ECM are called scaffolds,
and tissue engineering is the development of these artificial
scaffolds. Laboratory-grown tissues, cells and/or molecules are
cultured in a temporary three-dimensional scaffold to form the new
organ or tissue. The function of a scaffold is to act as a guide
and to direct the growth of the cells that migrate from the
surrounding tissue or the cells that have been seeded within the
scaffold prior to implantation. This method provides opportunities
to solve the organ donor deficiency problem. The demands of the
scaffolds are many; they must provide a suitable substrate for cell
attachment, proliferation, and cell migration. In addition, there
are several design criteria:
-
Introduction
4
− The material should have the correct pore size, pore
orientation, porosity and fiber structure
− The surface should permit cell adhesion
− The scaffolds should be biocompatible and degradable
− The material should maintain its form stability, be
reproducibly processable into a three-dimensional structure and
mechanically strong
Resorbable polymers are preferred in medical applications,
because polymers that do not degrade carry the permanent risk of
giving unwanted tissue responses.2 It is also advantageous in
self-repair processes to have a device that can be used as an
implant but does not require a second surgical intervention for
removal. Polymers, both degradable and non-degradable, are already
used in the body today to assist and replace the function of organs
and tissues.3 The applications using "biomedical" polymers range
from the long-term, as with a pacemaker casing, to the short-term
like a suture.4,5 Because of the wide spectra of applications the
rate and extent of degradability of a polymeric biomaterial must be
predetermined for each assigned function. Factors influencing the
degradability are, for example, chemical structure, copolymer
composition, architecture, molecular weight, morphology, surface
area and medium character.6 Tailoring an implant for controlled
degradation and transfer of stress to the surrounding tissue as it
heals at an appropriate rate is one of the greatest challenges
facing researchers today.
2.2 Monomers
Among the various families of degradable polymers, aliphatic
polyesters have a leading position. They are most effectively
derived from ring-opening polymerization and they have long been
considered as degradable materials for medical applications.7-10
The interest has been high since the hydrolytic and/or enzymatic
chain cleavage yields ω-hydroxyacids, which are in most cases
ultimately metabolized.
2.2.1 L-Lactide (LLA)
O
O O
OOH
O
O n O
O
H1
23
4
56
12
34
56
L-lactide(dimer of lactic acid)
poly(L-lactide)
Figure 2.1 Structure and properties of L-lactide (LLA) and
poly(L-lactide) (PLLA).
- semicrystalline - Tm PLLA 170-190˚C - Tm LLA 97˚C - Tg PLLA
55-60˚C
-
Introduction
5
Poly(L-lactide) (PLLA), Figure 2.1, is a semicrystalline polymer
able to form spherulites and lamellar crystals.11 The polymer is
degradable having low immunogenicity. It is considered to be
biocompatible and is often utilized as a medical material.12-14
PLLA belongs to the group of poly(α-hydroxy acids), and the
hydrolysis of PLLA yields lactic acid. Lactic acid is also a
by-product of the anaerobic metabolism in the human body. It is
incorporated into the tricarboxylic acid cycle and finally excreted
by the body as carbon dioxide and water. PLLA exhibits high tensile
strength and low elongation, and consequently it has a high modulus
making it suitable for load-bearing applications. However, PLLA is
a hydrophobic polymer having no reactive chain group and the use of
PLLA is therefore limited. It is difficult to chemically attach
active molecules like drugs and recognition agents onto these and
other polyesters. The rather low hydrophilicity, due to its
non-polar pendant methyl substituents, results in a limited water
uptake. This in turn results in a slow hydrolytic degradation rate
and the long-term biocompatibility can be affected. The degradation
kinetics of the implant are important for its biocompatibility. It
is well accepted that the degradation by-products are responsible
for tissue reactions and if large quantities of by-products are
released per time unit, they cannot be adequately handled by the
clearing capacity of the surrounding tissue.15-17
2.2.2 ε-Caprolactone (ε-CL)
O
O
O CH2 n H
n
O
C OH5
12
3
4 5
6
7
7 1
poly(ε-caprolactone)ε-caprolactone(lactone = inner ester) Figure
2.2 Structure and properties of ε-caprolactone (ε-CL) and
poly(ε-caprolactone) (PCL).
The caprolactone monomer, Figure 2.2, is a colorless liquid.
Poly(ε-caprolactone) (PCL) is a partitially crystalline degradable
thermoplastic polymer with ε-hydroxycaproic acid as the major
degradation product. The field of application is wide and includes,
for example, resins for surface coatings, adhesives and fabrics. It
also finds a use as stiffener for orthopedic splints, compostable
bags and sutures.18 The degradation rate is slower than that of
PLLA, and it is designed for use in long-term implantable systems.
Table 2.1 compares the mechanical properties of PLLA and PCL.
Table 2.1 Comparison of the mechanical properties of commercial
PLLA and PCL.19
Polymer Tensile strength[Pa ⋅10-5]
Elongation(%)
Modulus [Pa⋅10-5]
PLLA 500-800 5 - 10 27-40 ⋅103 PCL 200-300 300 - 500 2-3
⋅103
- semicrystalline - Tm PCL 58-65°C - Tg PCL -60ºC
-
Introduction
6
2.2.3 1,5-Dioxepan-2-one (DXO)
O
O
O
HO
O n
O
OH
12
3
4 5
6
7
1
7
6
5
4
3
2
Figure 2.3 Structure and properties of 1,5-dioxepan-2-one (DXO)
and poly(1,5-dioxepan-2-one) (PDXO).
The synthetic route to obtain 1,5-dioxepan-2-one (DXO) was
published for the first time in 1972.20 Poly(1,5-dioxepan-2-one)
(PDXO) is an amorphous polymer, and amorphous thermoplastics
without chemical and physical crosslinks do not have any form
stability, which is a drawback in many applications. PDXO
copolymers of different kinds have therefore been
investigated.21-23 Copolymerization with PLLA has the advantage
that the crystallinity, brittleness and high melting point of PLLA
is decreased. The copolymers show a low stiffness and high
elasticity compared to PLLA. The DXO/LA copolymers are interesting
materials, with possible applications in e.g. the biomedical field.
The degradation has been studied and the copolymer is hydrolyzed
mainly by ester bond cleavage.24, 25 Degradation studies of
triblock copolymers, PLLA-PDXO-PLLA, revealed that the degradation
rate was influenced by the original molecular weight and not by the
composition.26 Microspheres of the copolymers have been prepared
and the drug release pattern investigated, by altering the
components in the polymer could the degradation and erosion be
varied.27 Another alternative way of improving the stability of
PDXO is by chemical crosslinking, either by using tetrafunctional
bis(ε-caprolactone) as a crosslinking agent or by
photocrosslinking.28, 29 The crosslinked films have a high degree
of swelling in chloroform, are elastic without crystallinity and
have a high glass transition temperature (Tg). Crosslinked PDXO has
been used as substrate for grafting acrylamide in attempts to
design new degradable systems.30
2.3 Ring-opening polymerization
Aliphatic polyesters can be synthesized through either
polycondensation of acids and alcohols or ring-opening
polymerization (ROP) of cyclic esters. In contrast to the
traditional step-polycondensation method, the ROP of a cyclic ester
is an effective method of preparation of an aliphatic polyester.
Under rather mild conditions, high molecular weight aliphatic
polyesters can be prepared in short periods of time.8, 31, 32 The
development of ROP of lactones, anhydrides and carbonates started
around 1930.33-36
ROP of lactones can be carried out in the melt, in the bulk or
in solution and by, for example, cationic, anionic, free radical,
active enzymatic or coordination-insertion mechanisms depending on
the monomer and the catalyst.37 During a typical ROP
- amorphous - Tg PDXO -39˚C
-
Introduction
7
reaction, the chain end reacts with a new monomer during the
propagation step and the kinetics during the polymerizations then
follow the typical pattern of a chain-growth polymerization.
Numerous publications during the past years have shown that it
is possible to use ROP in the living and controlled polymerizations
of cyclic esters.32, 38, 39 Living polymerization means that the
initiated species maintain their activity until all monomers have
reacted. There is no irreversible deactivation (termination) or
irreversible transfer. Kinetic studies can be carried out to
elucidate whether the reaction is living. If there is no
termination during the polymerization,
( )ion]concentrat ln[monomer[M][M]
ln-0
t −== should be a linear function of time. Where
[M]0 is the initial monomer concentration and [M]t the
concentration at a given reaction time. Without any transfer, the
degree of polymerization (DP) should be a linear function of
monomer conversion. It is correct to say that the polymerization is
living if both these plots are linear.40
2.3.1 Coordination-insertion mechanism
When metal alkoxides containing free p- or d-orbitals of a
favorable energy (Mg, Sn, Ti, Zr, Fe, Zn, Al, Sm, Zn-alkoxides) are
used as initiators, a ”coordination-insertion” mechanism is
proposed, Figure 2.4.38, 41
− Lactone complexation to the initiator
− Monomer insertion into the metal-oxygen bond
M OR
OR'
O
MOR
OR'
O
M O R'
O
OR+coordination insertion
Figure 2.4 The proposed reaction pathway of
coordination-insertion ROP of cyclic esters.
When the metal atom of the alkoxide contains a free p or d
orbital, which means that the orbital contains no electron, the
metal atom attaches to the oxygen of, for example, lactide and the
-OR part of the alkoxide attaches to the C=O of the lactide. Since
the oxygen in the lactide molecule has a lone pair of electrons,
this lone pair has to be shared with the metal atom. This sharing
can occur only when the metal contains free p or d orbitals.
However, free p or d orbitals alone would not greatly assist
sharing; they have to be of similar energy. The energy level has to
be appropriate for the electron of the oxygen to jump to that
energy level of the metal. This mechanism involves a rearrangement
of polarized covalences and no ionic species.42-45 This is an
active research area where new compounds are continuously being
tested as initiators.
-
Introduction
8
2.3.2 Ring-opening polycondensation
Ring-opening polycondensation reactions can be used for the
synthesis of new molecular architectures. One example is presented
in Scheme 2.1, were the initiator is cyclic and has two reactive
bonds. The different functional groups can react with each other
and the kinetics follows the step-growth polymerization. Polymers
with different end groups as well as networks have been synthesized
by this method.46, 47
Scheme 2.1 Schematic presentation of ring-opening
polycondensation.
Bu2SnOO
CH2 n ClC X CCl
OO-Bu2SnCl2
O CH2 On X C
O
C
O
xl
Bu2SnOO
CH2 n
ClC X CClOO
+
2.4 Macromolecular design
ROP enables the polymer molecular weight and backbone
stereochemistry to be controlled and it can yield macromolecular
samples with narrow molecular weight distributions. Such a tuning
of polymerization reactions is important because of the close
relationship between molecular characteristics and material
properties. A representative example of the correlation between
structure and properties is the differences between PCL and
2-oxepan-1,5-dione. PCL cannot be used as a packaging material
because its melting temperature (Tm) is too low. By synthesizing
2-oxepane-1,5-dione instead, which has the same structure as CL
except that the central methylene group is replaced by a carbonyl
group, a semicrystalline polymer with a high Tm (147°C) is
obtained.48
2.4.1 Copolymerization
The idea behind a resorbable material in medical applications is
that the scaffold should act as a temporary replacement while the
tissue is regenerating. A major problem is to find a material that
starts to degrade and lose its mechanical properties in a
predetermined way while the damaged tissue is regenerating and then
disappears afterwards without a trace. The material needs good
mechanical properties to be able to stand up to all the complex
forces and maintain its form stability, and for this reason
crystalline materials are often used. The drawback with crystalline
materials is that they can cause adverse tissue responses. The
material has hard edges, which cause
-
Introduction
9
irritation and inflammation. A crystalline material also becomes
brittle during degradation, the surface cracks and pieces from the
surface are loosened.49-51 Amorphous polymers are much better
materials in this sense, but instead they may suffer from poor
mechanical properties. To meet all the demands, degradable block
copolymers have been found to be promising biomaterials because of
the potential to manipulate their amphiphilic behavior, and their
mechanical and physical properties by adjusting the ratio of the
building blocks or by adding new blocks of desired properties. This
is one of the simplest and most widely used ways of modifying
polymer properties to meet specific requirements, and it involves
random, alternating copolymers and segmental block copolymers. The
research in this area has been going on for many years and there
are many publications regarding block copolymers and their
properties.
Efforts to increase and vary the hydrophilicity of PLLA using
copolymerization are well documented because PLLA is often used as
a polymer in medical applications and its high hydrophobicity is a
limiting factor. Changing the hydrophilicity also changes the
degradation rate and profile of the PLLA. New possibilities for the
drug delivery industry and others are created in cases where both
the hydrophilicity and degradation profile are controlled.
Copolymerization can also be used to solve the second problem
associated with PLLA, namely the absence of reactive sites. Sites
which can be tuned and that allow the selective attachment of
substances like bioactive molecules have been synthesized by
copolymerization.52 Most interesting in this context is the
copolymerization of PLLA and poly(ethylene glycol) (PEG). There are
numerous possibilities to influence the properties and optimize
them for medical applications. It has been shown that adjusting the
block lengths of the components can modulate the crystallinity and
that the hydrophilicity is improved compared with that of the PLLA
homopolymers.53-56 The melting points of PLLA-PEG copolymers are
lower than that of PLLA and, with the incorporation of PEG as the
center block in the PLLA homopolymer, the elasticity and toughness
of the resultant copolymer are higher than those of PLLA.57
Cytotoxicity tests for the triblock have been performed, and these
showed a high level of cytobiocompatibility.58 Positive results
regarding micro-domain structure and drug-release properties using
block copolymers of PLLA and poly(ethylene oxide) have also been
presented.59, 60 An interesting system in this context are also the
block copolymers of PLLA and PDXO. PDXO is an amorphous polymer
and, even in this case, the hydrophilicity and mechanical
properties of the polymer can be tuned by using different
compositions of the monomers.61, 62
2.4.2 Functionalized macromonomers
Another way of synthesizing new advanced and controlled
molecular structures with specific properties is by using
macromonomers.63-65 Depending on the functional group and
modification method, different kinds of architectures can be
obtained, Figure 2.5. All architectures will possess unique
properties.
-
Introduction
10
Figure 2.5 Schematic representation of a) ladder b) comb-shaped
polymer c) star-shaped homopolymer d) star-shaped block copolymer
e) cyclic polymer f) network.
The functional groups can be inserted into the main chain during
polymerization or already into the monomer.48, 66-69 One example of
the usefulness of functionalized molecules is to be found in the
brominated polyesters.70 The bromide can be converted into an
unsaturated group using tertiary amines or dehydrohalogenating
reagents. The unsaturated bonds make the polymer suitable for
crosslinking, and this is useful for the synthesis of degradable
networks. The unsaturated units can also be converted into other
functional groups, such as epoxy, carboxylic acid, and hydroxyl
groups. Epoxidation with, for example, peroxy acids is one of the
most important reactions for the introduction of oxygen atoms into
organic molecules and is often used in syntheses.71-75 The polymers
can also be functionalized during polymerization. For example,
acrylic macromonomers of PLLA have been synthesized using
functionalized aluminum alkoxides as initiators.76 The
macromonomers obtained are suitable for graft copolymerization.
Basically, for the synthesis of graft copolymers, the macromonomers
can be end-functionalized (grafting onto), and the macromonomer
backbone can be functionalized (grafting from). The three main
strategies commonly used are:
1) Copolymerization of a macromonomer with vinyl or acrylic
comonomers
-
Introduction
11
2) Grafting-onto
3) Grafting-from
Each option can be used in copolymerization in order to achieve
the desired properties.
The material properties can be varied and in some cases even
controlled by changing the architecture. The thermal properties
will of course be affected when branching points are introduced
into the polymer. The chain length will decrease and the number of
chain ends increase. Both these give a lower melting point because
of the less ordered fold pattern of the crystal. It has been
concluded with three- and four-armed star-shaped PCL that it is the
arm length that affects the melting point rather than the total
weight.77 The lower melting points and lower melt viscosities will
be a major advantage for the melt processing of polylactides for
e.g. sutures. Long-chain branches predominantly affect the
viscoelasticity, decreasing the viscosity and increasing the
elasticity, and short-chain branches mostly affect the
crystallinity.78 This offers possibilities to adjust the
crystallinity by variation of the chain length and their number,
and this can be utilized and optimized for each application. By
manipulating parameters such as chain length, composition and
molecular weight, both the three-dimensional structure and the
hydrophilicity of polyesters can be varied. Grafting short
hydrophobic PLA chains to a hydrophilic backbone generates
polyesters with a more rapid water uptake and faster degradation
rates. Since aliphatic polyesters are thought to degrade by random
hydrolytic cleavage of the ester bonds, crystallinity and water
uptake are the key factors determining the rate of polymer
degradation. The degradation rate can thus be controlled by not
only the by crystallinity but also by the molecular
architecture.78, 79
-
Introduction
12
2.4.3 Star-shaped polymers
Star-shaped polymers are branched polymers with more than two
linear polymeric arms attached to a central core. These polymers
provide a lot of end groups which are used in subsequent
derivatization in, for example, surface functionalization.
Star-shaped degradable aliphatic polyesters have a great potential
in biomedical applications due to their high polymer mass and high
functionality per unit volume. They are also interesting since
using different number of arms varies the physical properties as
well as the degradation rates and the properties are different from
those of linear polymers.80, 81 There is a challenge in producing
well-defined stars in terms of the dispersity of arm number and arm
length. Among various attempt to synthesize polyesters with this
architecture, there are two methods that are most often used and
can be classified and distinguished. The methods are similar to the
ones described in the last section:
− Core first, living polymerization with a multifunctional
initiator 82, 83
− Arm first, coupling reaction of linear living polymers with a
multifunctional coupling agent 84
The method most often described in the literature for the
synthesis of star-shaped polyesters is one in which stannous
octoate (SnOct2) is used as a catalyst together with a
multifunctional alcohol. SnOct2 is widely used because it is
commercially available, soluble in common organic solvents and
cyclic ester monomers, and a permitted food additive in numerous
countries. The drawback is that it is difficult to achieve the
architecture without any imperfections.85 There is a need for new
methods for the synthesis of well-defined star-shaped polymer.
Kricheldorf and coworkers were the first to introduce spirocyclic
tin initiators, which were supposed to be used to synthesize
star-shaped polymers and networks in a precise way.86 The initiator
was synthesized from dibutyltin oxide and pentaerythritol, but it
did not fulfil all expectations. The initiator had a low solubility
in organic solvents and it was difficult to use. The continuation
of this work showed promising results with well-defined structures
as a result.47
In this connection, the hyperbranched and dendritic
architectures must also be mentioned. The polymers are highly
branched, resulting in many end-groups and an amorphous material
with low melt-viscosity and high solubility. There has been a
tremendous development in this area during recent years, and
controlled architectures can now be synthesized and the
conformation of the polymer can be controlled by
self-assembly.87-89
2.5 Networks
Another class of interesting synthetic materials with many
possible applications is that of polymer networks in which several
linear polymer chains are interconnected.
-
Introduction
13
Aliphatic polyesters sometimes present strength limitations and
the use of networks represents one way of providing the necessary
strength and rigidity. Different methods have been used to
construct such networks.75, 84, 90-92 In many of these methods, it
is not possible to control the chain lengths between the crosslinks
of the polymer network. If the network cannot be synthesized in a
controlled and predetermined manner, the properties are difficult
to tailor. The molecular structure of the crosslinked polymers has
been shown to affect the dynamic mechanical and swelling
properties.93-96 The crosslinking of homo- and copolymers provides
further possibilities for modifying the physical and mechanical
properties of materials.97
2.5.1 Hydrogels
Hydrogels, water-swellable networks, are three-dimensional,
hydrophilic, polymeric networks able to capture large amounts of
water. The network is composed of homopolymers or copolymers, and
is insoluble due to the presence of chemical crosslinks or physical
crosslinks, such as entanglements or crystallites.98 Hydrogels are
useful as biocompatible synthetic materials, especially in short-
and intermediate-term applications.99, 100 They are used as
super-absorbents, tissue-engineering scaffolds, sensors, chemical
memories, molecular separation systems, drug delivery systems and
other biomaterials.101-105 Hydrogels are preferred and are useful
in medical applications because of their similarity to natural
living tissue due to their high water contents and soft
consistency. The properties that make hydrogels useful in medical
applications are:
− The hydrodynamic properties of the hydrogels are similar to
those of the tissue
− The frictional irritation due to the presence of the hydrogel
is low because of its soft and rubbery nature
Polymer hydrogels containing both hydrophobic and hydrophilic
units are called amphiphilic polymer hydrogels. Compared to simple
homopolymer hydrogels, they have improved mechanical properties
because of the presence of the hydrophobic units. This reduces the
water content and produces a more coherent material. The presence
of both hydrophilic and hydrophobic segments enables the materials
to be used in the release of both hydrophilic and hydrophobic
drugs.106 The side-chain length, degree of crosslinking, swelling
kinetics and composition strongly affect the release behavior and
other mechanical properties of the hydrogel.107-109 For example,
when the hydrodynamic radius of the solute is much larger than the
mesh size of the network, the solute-release behavior is controlled
by the degradation.110 Another growing field were hydrogels are
preferred is that of molecular imprinting. Hydrogels can bind
analytes and also choose between different molecules.111 In the
future, an exact and well-defined network structure seems to be
necessary. It is therefore important to develop methods for the
synthesis of well-defined hydrogels.
-
Introduction
14
2.6 Biomedically adapted surfaces
Integrins are cell-surface receptors that mediate adhesion to
the ECM proteins, Figure 2.6. Most cells use several integrins that
recognize a range of ECM associated ligands. The integrins play
important roles in differentiation and cell communication.
Figure 2.6 Schematic picture of cell adhesion to the ECM
proteins through integrins.
Synthetic polymers have no natural cell binding sites, the cell
adhesion occurs instead via proteins.112 When a material is
implanted into the body, the material surface is exposed to the
proteins present in blood and other body fluids. This results in a
layer of proteins adsorbed to the surface. The proteins compete for
the surface, since they have different affinities.113 The material
surface smoothness, ionic and electronic charge, wettability by
blood components and chemical structure determine the compositions
of the proteins and bacteria that adhere to the surface.114-116
This in turn decides how the cells respond to the material surface.
The protein layer is also the beginning of a vascular fibrous
capsule, which will affect the adhesion of platelets and also
influence other coagulation processes. It would be best if the
implanted devices exhibited a “normal” wound healing. A normal
wound healing is regulated by growth factors. These affect the
protein adhesion and thereby the cell migration and proliferation.
A normal wound healing after implantation would provide an
integration of the implant with the body and not segregation
through a vascular fibrous capsule. Therefore, ways to hinder the
vascular fibrous capsule from being built are currently being
sought. To improve the cell affinity, many efforts have been
directed towards modifying the surface properties for example by
adjusting the hydrophilicity, hydrophobicity and surface
roughness.117-121
2.6.1 Cell adhesion – hydrophilicity and hydrophobicity
Surface hydrophilicity plays an important role in cell adhesion,
spreading and growth. This may be because the proteins that adhere
prefer the hydrophilic surface. Hydrophobic surfaces have a high
interfacial free energy in aqueous solutions and this
-
Introduction
15
seems to be a disadvantage in terms of their cell, tissue, and
blood compatibility. Surface modification to achieve a more
hydrophilic polymer is the most common way of altering the surface
of a particle so that it can be ingested by phagocytes and of
improving the surface properties of the system. This is still a
complex area which is far from being well understood. Articles
describing how cells adhere less to hydrophilic surfaces have also
been published.122, 123
2.6.2 Cell adhesion - morphology and topography
The key design parameter for achieving good cell responses is
the sample shape and the nano-level topography of the
material.124-126 The topography of natural soft tissue is dependent
on the ECM, which have a nanometer length and width. Successful
polymeric constructional materials should therefore have
nano-dimensional surface features. It is expected that a
biocompatible material that mimics the nanometer topography of the
relevant tissue will enhance cellular responses, and thus lead to
better tissue integration in vivo. The interaction between the
scaffold and cell also determine the cell function. It was shown
that not only proliferation but also cell differentiation and cell
migration depended on interactions with polymer surfaces.122 The
cells recognize surface features and react to them, resulting in
some kind of contact guidance. This can be used in different ways.
Ligaments and tendons are well-organized fibrous connective tissues
but, after an injury, cells in the healing site are found to have a
non-specific orientation. The resulting collagen matrix is also
less organized. The explanation of the decrease in mechanical
properties of the healing tissue has been related to these
unorganized structures. When the broken ligaments and tendons are
treated it is important that the cells are aligned and that the
collagen matrix is organized as in normal tissue. The topography of
a surface can help the cells align. Orientation of the cells and
also an organization of the collagenous matrix was achieved by
using a structured membrane. The cells then produce aligned
collagenous material similar to the uninjured state of tendons and
ligaments.127 Research has also been carried out to elucidate
whether a surface with nanometer feature dimensionalities has any
effect on the cell adhesion. The results show a clear effect, the
numbers of cells were compared to a flat surface and the number of
adhered cells was 51% higher on the nano-structured surface after 5
days.120 Other reports support this claim and provide evidence that
surfaces with smaller features enhance cell functions.3, 121,
125
The molecular architecture of the polymer is of course a major
factor during the micro-phase separation and for the topography
obtained.128, 129 The microphase-separated morphologies also have a
pronounced effect on the mechanical properties.130, 131 The modulus
is dependent on the fraction of taut interfibrillar tie molecules,
while the tensile strength depends on the ratio of interfibrillar
area to total area.
-
Experimental
17
3. EXPERIMENTAL 3.1 Materials
The L-lactide (Serva Feinbiochemica, Germany, 98%) was
recrystallized from toluene several times, dried at room
temperature under vacuum for 48h and stored in an inert atmosphere.
ε-caprolactone (Acros) was dried over calcium hydride for 48h at
room temperature and distilled under reduced pressure just before
use. The synthesis of 1,5-dioxepan-2-one (DXO) has been described
elsewhere.132, 133 After the synthesis DXO was purified by two
distillations, recrystallization from dry diethyl ether and a final
destillation under reduced pressure. m-chloroperoxybenzoic acid
(mCPBA; Aldrich) was cleaned from m-chlorobenzoic acid by
dissolving in dichloromethane and washing with a phosphate buffer
at pH 7,4. The organic layer was dried over MgSO4, filtered and
dried under a vacuum for 2 days at room temperature. Toluene
(Merck, Germany) was dried over Na-wire before use. Chloroform
(Labora Chemicon, Sweden) stabilized with 2-methyl-2-butene was
dried over calcium hydride for at least 24 h and then distilled
under reduced pressure under an inert atmosphere. Dibutyltin oxide
(Aldrich, Germany), Succinyl chloride (95%, Acros Organics),
Dichloromethane (VWR) and pentaerythritol ethoxylate, 3/4 EO/OH and
15/4 EO/OH, (Aldrich, Sweden) were used as received.
3.2 Synthesis of initiators 3.2.1 Germanium initiators
GeO
O
O
O
O
n
OO
n
OO
O
Figure 3.1 Germanium initiators, 1 (n=4), 2 (n=11) and 3
(n=43).
The germanium initiators, 1, 2, and 3, were a gift from
Professor Kricheldorf and their synthesis was not a part of this
work. Appearance: 1 had a syrupy character while 2 and 3, had a
more crystalline shape, Tm (2) 72°C, Tm (3) 65 °C.
-
Experimental
18
3.2.2 Functionalized tin initiators
Figure 3.2 Functionalized tin initiators, 4 and 5.
The functionalized initiators 4 and 5 were synthesized from
dibutyltin dimethoxide and the corresponding alcohols as described
in the literature.134, 135 Initiator 5 was recrystallized in dry
toluene, and both initiators were distilled over a short-path
apparatus under reduced pressure (10-3 mbar) before use.
Appearance: 4 had a syrupy character and 5 was crystalline, Tm (5)
90°C.
3.2.3 Spirocyclic tin initiators
SnSnO
Op
OO o
O
O
On
Om
Figure 3.3 Spirocyclic tin initiators, 8 (m+n+o+p=3) and 9
(m+n+o+p=15)
The procedure was the same as in the synthesis of the
functionalized initiators. Two different pentaerythritol ethoxylate
compounds were used. The precipitated product was centrifuged
before the supernatant solvent was poured off. The initiator was
dried under reduced pressure for 24 h. Elemental analysis (%): 8
C27H56O7Sn2 (731.8) calculated C 44.3 H 7.7, found C 42.4 H 8.0, 9
C51H104O19Sn2 (1258.8) calculated C 48.6 H 8.3, found C 46.6 H 8.2.
Exposure to moisture before analysis could explain the difference
between the calculated and found amounts of carbon. Appearance:
crystalline solids, white, Tm (8) 117°C Tm (9) 109°C
3.3 Polymerization model reaction
The polymerizations were carried out in silanized round-bottomed
flasks closed by a three-way valve. A magnetic stirring bar was
enclosed in the reaction flask. The
SnO
OSnSn
O
O
O
O2
SnO
O
OOSn
O
Sn
O
2 4
5
-
Experimental
19
equipment was flamed and stored in a glovebox (Mbraun MB
150B-G-I) purged with nitrogen. The monomer and the initiator were
weighed and added to the reaction vessel in the glovebox. Distilled
chloroform was transferred to the reaction vessel in the hood by a
flamed syringe under strictly anhydrous conditions. During
polymerization, the reaction flask was completely immersed in a
thermostated oil bath preheated to the polymerization temperature.
Samples for 1H-NMR and SEC analysis were withdrawn from the
reaction vessel using a flamed syringe while flushing with inert
gas. The product was precipitated in cold methanol/hexane mixture
when the reaction time was over.
3.4 Epoxidation
Epoxidation was carried out in chloroform. To obtain the
completely epoxidized products the amount of mCPBA used in the
epoxidation reaction was set to twice the theoretical number of
double bonds. mCPBA was added to a round-bottomed flask containing
PLLA dissolved in chloroform. The reactions were maintained with
magnetic stirring at room temperature until the conversion of
double bonds was complete. A white precipitate appeared and, after
filtration, the filtrate was precipitated in cold hexane to obtain
the epoxidized polymer.
3.5 Copolymerization
The first step, polymerization of DXO, followed the procedure
described in the model reaction, section 3.3. During the
polymerization, the flask was immersed in a thermostated oil bath
at 60°C. The initial monomer concentration was 0.5 M. The second
monomer, LLA, was dissolved in chloroform following the same
procedure as for DXO and transferred to the reaction vessel with a
syringe at the time for full conversion of DXO.
3.6 Synthesis of networks 3.6.1 Tetra-functional acid
chloride
The synthesis was performed according to an earlier report.136
Cis-1,2,3,4-cyclopentane tetracarboxylic acid and n-heptane were
added to a round-bottomed flask containing phosphorus
pentachloride. The temperature was increased gradually from 20°C to
95°C over a three-hour period. Reflux was maintained at 95°C for an
additional 2 hours until no HCl evolution was detected. The yellow
solution obtained was filtered through a coarse paper and then
roto-evaporated to oil.
3.6.2 Crosslinking reaction
When the polymerization was complete, the crosslinker was added
through a flamed syringe. In most cases a gel was formed
immediately as soon as the crosslinker
-
Experimental
20
was added. The reaction was held at 60°C for an additional
couple of hours to ensure complete conversion of the acid chloride.
The gels were extracted with CH2Cl2 before characterization.
3.7 Film preparation
The polymer was dissolved in chloroform to form solutions with
concentrations of 0.3 wt% and 5 wt%. A freshly cleaved mica
substrate (9 cm × 10 cm) was put into a glass container and 2 ml of
the solution was deposited on the surface of the substrate. The
samples were conditioned for 2 days at room temperature and then
for at least 1h in vacuum. The samples were heated in vacuum at
170°C for 16h and then either quickly quenched to room temperature
or slowly cooled to room temperature over a period of several
hours.
3.8 Characterization methods 3.8.1 Nuclear Magnetic
Resonance
For 1H-NMR measurements, the samples were dissolved in
deutero-chloroform in 5-mm NMR tubes at room temperature. The
sample concentration was 5% by weight. 13C-NMR was performed with a
10% sample concentration in 5-mm tubes. Non-deuterated chloroform
was used as an internal standard (δ = 7.26 ppm). NMR spectra were
recorded on a Bruker AM-400 Fourier-Transform Nuclear Magnetic
Resonance spectrometer (FT-NMR) operating at 400 MHz, T=25°C. When
spectra of the functionalized initiators were recorded at different
temperatures a Bruker DMX 500 was used. 2D 1H- 13C heteronuclear
multiple quantum coherence – gradient selected (invieagssi) spectra
were acquired and processed with a standard Bruker microprogram. A
total of 256 experiments were accumulated using one scan with a
relaxation delay of 2s. The spectrum was obtained with 9 ppm
spectral width over the F2 (proton) axis and 200 ppm for 13C along
the F1 (carbon) axis at -13°C.
3.8.2 Size Exclusion Characterization
Chloroform was used as an eluent and was delivered at a flow
rate of 1.0 mL/min. The samples were dissolved in chloroform at a
concentration of 0.06 wt%. The injection volume was 50µL. Narrow
polystyrene standards in the 580-1,900,000 g/mol range were used
for calibration. A Waters 717plus auto sampler and a Waters model
510 apparatus equipped with three PLgel 10 µm mixed-B columns,
300×7.5 mm (Polymer Labs., UK) connected to an IBM-compatible PC
were used. Millenium32 version 3.20 software was used to process
the data.
-
Experimental
21
3.8.3 Differential Scanning Calorimetry
Differential scanning calorimetry (DSC) measurements were made
on a Mettler-Toledo DSC instrument with a DSC 820 module. The
measurements were run from -50°C to 180°C at a heating rate of
10°C/min and a cooling rate of 10°C/min. The samples were heated in
a nitrogen atmosphere. Tg and Tm of the polymers were determined
during the second heating period. Tg was determined as the middle
of the record step change in heat capacity, and the Tm was defined
as the endotherm peak of the curve.
3.8.4 Atomic Force Microscopy
The atomic force microscopy (AFM) measurements in the tapping
mode were made on a Multimode Instrument from Digital Instruments
equipped with a Nanoscope III software system. Commercial etched
silicon nitride cantilevers of 125 µm length with a spring constant
of 36 - 55 N/m and a resonance frequency of 324 - 372 kHz were
used.
3.8.5. Environmental Scanning Electron Microscopy
ESEM (Environmental Scanning Electron Microscope) model 2020
produced by ElectroScan. Thermoelectric stage was used in the
experiments, which alters and measures the temperature. The wetting
conditions were created by maintaining the temperature at 4°C and
by altering the pressure in sample chamber from 3.0 to 7.0 Torr.
The water was condensed from the chamber atmosphere.
3.8.6 Swelling
The swelling of the network was studied gravimetrically. All the
swelling data were obtained with extracted hydrogel specimens. In a
typical case, a piece of the network film was weighed and
transferred to water. At regular intervals, it was taken out, the
excess water was removed from the surface with tissue paper and it
was then weighed and returned to the medium. This procedure was
continued until constant weight was attained. The equilibrium
degree of swelling, DS, was calculated as:
degree of swelling (DS) ( )
0
0100W
WW −⋅= [3.1]
where Wo is the initial weight of the dry sample and W is the
final weight of the swollen sample. Each measurement was repeated
three times and the average value was reported.
-
Experimental
22
3.9 Cell response measurements
The group of Professor Biagini, Italy, made all the measurements
of cell growth.
Keratinocytes, NCTC 2544 cells, (ICLC Genos Italy) were grown on
the heat-treated samples in a controlled atmosphere (5% CO2;
T=37°C) in Minimum Essential Medium Eagle (MEM) (Sigma, Milan,
Italy) supplemented with 5% foetal calf serum (FCS), 1%
non-essential amino acids, 2.0 mM L-glutamine, and antibiotics.
After thawing, keratinocytes were routinely split 1:2 every 3-4
days and used between the 2nd and 4th passages. For SEM analysis
the cells were fixed in 2% glutaraldehyde in 0.1 M cacodylate
buffer (pH 7.4), post-fixed in 1% osmium tetroxide, dehydrated in
increasing ethanol concentrations, CPD-dried, mounted on aluminium
stubs and gold-sputtered.
3.9.1 Time-lapse videomicroscopy
Cells were seeded onto the heat-treated samples in 2ml
Hepes-modified E-MEM supplemented with 5% FCS, 2ml L-glutamine,
100ml U/ml penicillin, 100U/ml streptomycin and kept at 37°C. They
were observed under an inverted microscope (Nikon Eclipse TS-100)
equipped with a 10x objective and a colour CCD video camera (JVC
TK-C1381). Phase-contrast images of living cells were recorded
using a time-lapse VCR (Panasonic AG-TL700) and digitalized using a
video frame grabber card and dedicated software (Image-Pro Express,
Media Cybernetics).
-
Triblock copolymers
23
4.TRIBLOCK COPOLYMERS The polyether-polyester block copolymers
are an interesting class of biomaterial. Copolymers of, for
example, PLLA and PEG provide a large variety in terms of
mechanical properties and degradability. PEG is hydrophilic and
flexible and PLLA appears most interesting due to its
degradability. In addition, both PEG and PLLA have been accepted by
the U.S. Food and Drug Administration for internal use in the human
body. All these properties are valuable for biomedical applications
such as implanting devices, materials for tissue engineering and
cell scaffolds. By using copolymerization, the hydrophilicity of
PEG can be combined with the degradability of PLLA, and the high
crystallinity of the PLLA will decrease due to the flexibility in
PEG. Most advantageous is the ability to modulate the degradation
rate and hydrophilicity of the polymer by adjusting the ratio of
its hydrophilic and hydrophobic constituents.
4.1 Germanium initiators
Kricheldorf et al have earlier thoroughly investigated the
cyclization of oligo- and poly(ethylene glycol) with dibutyltin
dimethoxide and have also used the macrocycles in
polymerizations.137-139 The same group has continued the research
in this area and has recently synthesized the same kind of
structures but with tin being replaced as the metal atom by
germanium. The metal center plays an important role in ROP.43 The
electrophilicity of the initiator metal center is of the outmost
importance. The tendency toward metal-oxygen bond formation follows
the accessibility of the initiator’s lowest unoccupied molecular
orbital (LUMO). More electrophilic initiators polymerize cyclic
esters more rapidly.140 Secondly, the molecular weight dispersity
(MWD) of the polymer also depends on the metal.45 This can be
explained by the energy difference between the highest occupied
molecular orbital (HOMO) and LUMO. This energy difference decreases
in a group of the periodic table and is directly proportional to
the activation energy of the transesterification reactions.
Compared to tin, which is often used as an initiator in ROP, the
energy difference between HOMO and LUMO in germanium is higher,
Figure 4.1, the activation energy for transesterification is
therefore also higher. Since germanium is more electrophilic and
also has a higher activation energy for transesterification
reactions than tin it is interesting to see how these compounds
serve as initiators.141, 142
-
Triblock copolymers
24
Higher energy difference between HOMO and LUMO
Higher electrophilicity
Figure 4.1 Part of the periodic table showing how the
electrophilicity and the energy difference between HOMO and LUMO
varies.
Certain germanium compounds also have a low mammalian toxicity
and they exhibit a clear activity against certain bacteria.143
Professor Kricheldorf gave three germanium initiators to our
group and their structures are shown in Figure 4.2. Both the
synthesis and the mechanistic consideration of these compounds were
outside the scope of this work.
GeO
O
O
O
O
n
OO
n
OO
O
Figure 4.2 Structure of germanium initiators used in the
synthesis of triblock copolymers, 1 n = 4, 2 n = 11, 3 n = 43.
The initiators have different lengths of ethylene oxide units in
the structure. These will be incorporated into the polymer during
polymerization and when low monomer-to-initiator ratios are used
the polymer will end up in an ABA block structure.
The assignment of the initiators was done by 1H-NMR (Figure
4.3), 13C-NMR (Figure 4.4), 1H-1H COSY NMR (homonuclear
proton-proton correlation spectroscopy) (Figure 4.5) and 1H–13C
hmqc-gs spectra (heteronuclear multiple quantum coherence –
gradient selected) (Figure 4.6). Signals typical of the –O–CH2CH2–
group were observed in the 1H-NMR spectrum and the assignment can
be seen in Figure 4.3. The peak noted as "a" emerging at 3.61 ppm
originated from the –O–CH2– protons directly attached to the
germanium atom. The peak at 3.72 ppm was assigned to the protons
that are positioned next to these protons, –O–CH2–CH2-. In Figure
4.4 the 13C-NMR spectrum of initiator 1 is shown. All carbons could
be seen and the coupling between the protons and carbons was
recorded with a 1H–13C hmqc-gs spectrum (Figure 4.6). The coupling
between the protons can be seen in Figure 4.5.
Al Si P Ga Ge As In Sn Sb
-
Triblock copolymers
25
[ppm]
3.103.203.303.403.503.603.703.803.904.00
a
b
c
[ppm]
Figure 4.3 1H-NMR spectra of germanium initiator 1, 2 and 3.
Figure 4.4 13C-NMR spectrum of germanium initiator 1.
6161626263636464656566666767686869697070717172727373
a
b
c
initiator 1
initiator 2
O
OO
O
O
O
O
O
Ge*
*a bc
initiator 3
-
Triblock copolymers
26
Figure 4.5 1H-1H-COSY NMR spectrum of initiator 1.
F1=conventional 1H-NMR spectrum, F2=conventional 1H-NMR
spectrum
Figure 4.6 1H–13C hmqc-gs spectrum of initiator 1.
F1=conventional 13C-NMR spectrum, F2=conventional 1H-NMR
spectrum
-
Triblock copolymers
27
4.2 Polymerization
In solution polymerization, it is extremely important that the
initiator and monomer are completely soluble in the solvent.
Chloroform, 1,2-dichloroethane and chlorobenzene are good solvents,
in which initiators 1, 2 and 3 are soluble. Conversion tests showed
that chlorobenzene as solvent, a temperature of 120°C and an
initial monomer concentration of 1 M was a good option to achieve a
successful ring-expansion polymerization of LLA using these
germanium initiators. It should be noted that solution
polymerization using chloroform and 1,2-dichloroethane respectively
as solvent was first used in an attempt to carry out the reaction
at a lower temperature, but with no success. 1H-NMR spectrum of
initiator 1 and a spectrum of precipitated PLLA initiated by 1 are
assigned in Figure 4.7. Note that the signal from the protons
closest to the germanium at 3.61 ppm in the top spectrum was
shifted downfield to 4.3 ppm during polymerization. The
disappearance of the peak at 3.61 ppm proves that all four Ge-OCH2
bonds participated in the polymerization.
O
OO
O
O
O
O
O
O
O
OO
Ge n
n
O
OO
O
O
O
O
O
Ge
OO
O
O
O
OO
O
O
O
O
O
O
O
OO
n
n
O
OH OOH
c'
a' b'*
*
*
* ab
c
d'
precipitation
c'
a' b'
d'e'
2n
-
Triblock copolymers
28
Figure 4.7 1H-NMR spectrum of initiator 1 (upper), precipitated
PLLA initiated by initiator 1 (lower). [M]/[I]=50, [M]0=1M ,
T=120°C
Conversion tests were performed by withdrawing samples from the
reaction flask during polymerization. The samples were
characterized by 1H-NMR. Since the 1H-NMR signals of LLA and PLLA
are different, the conversion could easily be checked by 1H-NMR
spectroscopy.144 The results were used to construct a plot of
monomer conversion versus time, Figure 4.8. The conversion was high
in all polymerizations; the maximum conversions for the
polymerizations initiated by 1, 2 and 3 were 91%, 93% and 95%
respectively.
Figure 4.8 Conversion of LLA as a function of time. [M]/[I]=50,
[M]0=1M, T=120°C
0
20
40
60
80
100
0 5000 10000 15000 20000
time [min]
conv
ersi
on [%
]
initiator 3
initiator 2
initiator 1
3.23.43.63.84.04.24.44.64.85.05.25.4
d'
c'e'
a
b
c
a'
b'
[ppm]
-
Triblock copolymers
29
From the samples withdrawn from the flask, the semilogarithmic
plot, -ln([M]/[M]0) versus reaction time, could be obtained. Where
[M]0 is the initial LLA monomer concentration and [M] the LLA
concentration at a given reaction time. The linearity of the
relationship in Figure 4.9 attests that the polymerization kinetics
of LLA, [M]/[I] ratio of 50 at 120°C, initiated by these germanium
initiators are first order in monomer. The linearity also indicates
that the amount of termination reactions is low and that the number
of growing chains is constant.
Figure 4.9 Semilogarithmic plot of –ln([M]/[M]0) as a function
of time, polymerization of LLA using 1, 2 and 3 as initiators.
[M]/[I]=50, [M]0=1M, T=120°C
Plots of number-average molecular weight, Mn, versus conversion
also gave straight lines, Figure 4.10, which means that the
frequency of transesterification reactions during the
polymerizations was low.
Figure 4.10 Mn as a function of conversion during polymerization
of LLA using 1, 2 and 3 as initiators. Mn was determined by SEC
using narrow molecular weight polystyrene standards. [M]/[I]=50,
[M]0=1M, T=120°C
0,0
0,5
1,0
1,5
2,0
2,5
3,0
0 5000 10000 15000
time[min]
-ln([M
]/[M
] o)
initiator 3
initiator 2
initiator 1
0
5000
10000
15000
20000
25000
30000
0 20 40 60 80 100
conversion [%]
Mn
initiator 3initiator 2
initiator 1
-
Triblock copolymers
30
Considering the long reaction times, the MWD was found to be
narrow. Initiator 3 had the broadest dispersity, around 1.4 and
initiators 1 and 2 had a dispersity around 1.2. All the
above-mentioned results indicate that the concentration of active
centers remains constant and that the transesterification reactions
are low. The polymerizations were in other words controlled. For an
initiator to be useful it is also necessary that the molecular
weights of the polymer can be determined and can be predictable
from the molecular composition of the initial reaction mixture. Two
polymerizations were therefore performed with each initiator using
different targets of DP, DP=25 and DP=50. The reaction times could
be predicted from the conversion curves. The results are presented
in Table 4.1.
Table 4.1 Results from the polymerization of LLA in
chlorobenzene at 120°C, [M]0=1M .
a) monomer/initiator ratio added to the reaction flask b) degree
of polymerization found after precipitation c) conversion,
determined from crude reaction mixture d) number-average molecular
weight determined by SEC using narrow polystyrene standards
The DP after precipitation were determined by quantification of
the 1H-NMR spectrum, signals b', c' and d' in Figure 4.7 were used.
Polymers with narrow MWD and with DP close to the values expected
based upon monomer/initiator loading were produced.
Because of the low reaction rate, an attempt was made to
increase the reaction rate. PLLA with a DP of 50 was synthesized in
bulk, T= 120ºC, using initiator 2. After 20h the conversion was
85%, Mn was 5000 and the MWD was 1.4. An even higher reaction rate
was achieved when the reaction was performed at a temperature of
160°C. The conversion was 90% after 10h, Mn was 4000 and MWD 2.0.
The broader dispersity is a consequence of the high reaction
temperature, since this increases the amount of side reactions.
4.3 Thermal characteristics
The thermal properties of the PLLA-PEG-PLLA copolymers were
evaluated. Results of DSC measurements are shown in Table 4.2.
Initiator [M]/[I]a DPb Time (h)
Conv.c (%)
Yield (%)
Mn,SECd MWD
1 25 31 116 92 85 7 000 1.2 1 50 53 190 96 91 13 400 1.2 2 25 29
103 96 94 7 300 1.2 2 50 53 168 94 90 16 200 1.2 3 25 29 95 93 90
28 500 1.4 3 50 45 162 90 90 41 400 1.4
-
Triblock copolymers
31
0 20 40 60 80 100 120 140 160
temperature [°C]
endo
Table 4.2 Thermal properties of PLLA-PEG-PLLA triblock
copolymers.
a) degree of polymerization found after precipitation b) the
glass transition temperature of PLLA c) the crystallization
temperature of PLLA d) the melting temperature of PEG
Each chromatogram had two Tm that corresponded to the PEG and
PLLA melting transitions respectively. It is of interest to note
that the melting temperature of the PLLA-PEG-PLLA was shifted to a
lower temperature region (120-150˚C) than the typical Tm of normal
PLLAs with equivalent molecular weights (150-160˚C). The decrease
in Tm is due to the presence of the PEG unit. The incorporation of
PEG into the backbone of the polymer hinders the chain ordering of
the polymer, and leads to a decrease in crystallinity. The melting
point depression of the PLLA block with an increasing molar content
of PEG can clearly be seen in Figure 4.11.
Figure 4.11 DSC chromatogram of PLLA-PEG-PLLA triblock
copolymers initiated by germanium initiators. [M]/[I]=25
Another noticeable phenomenon is the multiple melting peaks in
each case, which indicates the dispersion of crystal thickness due
to the greater MWD of the polymer chains. Polymers initiated by
initiator 1 and 2 showed only a very small melting peak associated
with PEG. The melting peak of the PEG segments has a tendency to
disappear.145 The explanation of this observation is that the PLLA
hard segments are
Initiator DPa Tg b (°C)
TcPLLAc (°C)
Tm PEGd (°C)
1 31 39.5 88.6 54.0 1 53 46.2 92.8 51.3 2 29 - - 54.4
2 53 32.4 95.1 51.2
3 29 - - 55.7 3 45 - - 54.5
initiator 3 initiator 2 initiator 1
-
Triblock copolymers
32
the first to solidify upon cooling, which makes it hard for the
PEG segments to move so that the crystallization of PEG is thereby
blocked. This is most apparent in a block copolymer with a small
content of PEG. The observed exothermic peak at 90-95˚C is related
to the cold crystallization of the PLLA segments.146 No
crystallization was observed during the second heating when
initiator 3 was used. This is considered to be due to the
differences in PEG chain length and the low crystallizability of
low molar mass PLLA.
-
Functionalized polyesters
33
SnO
OSnSn
O
O
O
O2
5. FUNCTIONALIZED POLYESTERS In the introduction, the importance
of reaction sites in the polymer chain was emphasized. In order to
modify polymers by chemical reactions, they should contain reactive
groups either in the backbone or in side groups. A versatile
functional group is the C=C double bond. In this work,
functionality in the form of a double bond has been introduced into
the initiators and subsequently incorporated into the macromonomers
during the polymerization. Two different tin initiators have been
used, and in both cases a double bond has been introduced into the
main chain. Tin initiators were used since their reaction rates are
high and the knowledge concerning their reactions have been
thoroughly investigated and are well understood.
5.1 Functionalized cyclic tin(IV)alkoxides
The functionalized initiators were synthesized by the
condensation reaction between dibutyltin dimethoxide and the
appropriate alcohol. The driving force in the reaction is the gain
in entropy during cyclization, where many small molecules are
formed instead of one large. The macrocycles must of course be
strain-free. An additional stabilizing effect results from the
intramolecular donor-acceptor interactions between oxygen and tin.
Such interactions have been shown to occur.147 The cyclic tin
alkoxides in Figure 5.1 have previously been mentioned briefly in
the literature.148 Published 119Sn-NMR analysis data for these
initiators showed that the unimer form of initiator 4, noted as 4/1
in Figure 5.1, predominates at room temperature, whereas initiator
5 is present mainly as a dimer, 5/2. In the continuation of this
work, the initiators will be referred to only as 4 and 5.
Figure 5.1 Functionalized tin alkoxide initiators used in
polymerizations with polyesters.
4/1
SnO
O
OOSn
O
Sn
O
2
4/2
5/1 5/2
-
Functionalized polyesters
34
a’’ b’’
c e d f
T= 7°C a b’ b’’
T= -13°C
a’ a’’ b’ b’’
1.0 7.0 6.0 3.0 5.0 2.0 [ppm]
e c
d
T= 25°C 4.0
a b
f
1H-NMR spectra of initiator 4 at different temperatures, T= +25,
+7, -13°C, are shown in Figure 5.2. The peak appearing at 4.49 ppm
originated from the two –O–CH2 protons next to the double bond in
the initiator. This peak was broad at room temperature, but when
the temperature was lowered this broad peak split into two peaks.
To understand this behavior, a 1H–13C hmqc-gs spectrum was recorded
(not shown), from which it became clear that the splitting of the
signal originated from a fast equilibrium between the unimeric and
dimeric forms, 4/1 and 4/2 respectively. At room temperature, the
rate of exchange between the monomeric and dimeric forms was too
fast for the signals to be resolved.
Figure 5.2 1H-NMR spectra and assignment of initiator 4 at
different temperatures. T= +25°C, +7°C, -13°C
Other oligomers than the dimers are probably present to some
extent, causing the broadening of the peaks. The dimerization is
probably due to the favorable Sn–O donor-acceptor interactions.
b’ a’
SnO
O
OOSn
O
Sn
O
2
-
Functionalized polyesters
35
j f g i
c, c'
e h
b d
c, c'
a
[ppm]
The 1H-NMR spectrum (T=+25°C) of initiator 5 with assignments is
shown in Figure 5.3. Also with this initiator, spectra were
recorded at lower temperatures, +7°C and –13°C, but this lowering
of the temperature did not have any effect. The reason is that the
rate of exchange between the two forms is slow enough for the peaks
to be resolved at 25°C. Kricheldorf has reported the same
observation for similar cyclic tin alkoxides.149
Figure 5.3 1H-NMR spectra and assignment of initiator 5.
5.2 Synthesis of functionalized polyesters
The polymerizations were carried out in chloroform at 60ºC with
an initial monomer concentration of 0.5 M. The reaction rate was
monitored by 1H-NMR spectroscopy. Samples were withdrawn from the
reaction mixture at different times and analyzed until no further
changes in the conversion could be observed. As can be seen in
Figure 5.4, the reaction rates of the two functionalized tin
initiators when polymerizing LLA were similar. No induction period
could be seen and the conversions were almost complete. With ε-CL,
the reaction rate was faster. Conversion of ε-CL was complete
within 140 minutes, in contrast to LLA, which was only converted to
45% during the same period of time, indicating that ε-CL was more
reactive than LLA.
SnSnO
O
O
OSn
O O
5.0
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.5 6.0 6.5 7.0
j e
i
a
f,g h
b d c, c'
-
Functionalized polyesters
36
Figure 5.4 Conversion curves for polymerizations of LLA and ε-CL
using the two functionalized tin initiators, 4 and 5. [M]/[I]=50,
[M]0=0.5M, T=60°C
The spectra of crude PLLA initiated by 4 at two different times
during polymerization together with the spectrum of pure initiator
4 are shown in Figure 5.5.
Figure 5.5 1H-NMR spectrum of initiator 4 (at the bottom) and
spectra of two samples taken during polymerization of LLA.
In all the studied cases, the signal at 4.49 ppm associated with
the unreacted initiators disappeared. The resulting polymers should
contain the initiator fragment as one type of block. The peak from
the incorporated –CH2CH=CHCH2– group of initiator 4
3.54.04.55.05.5(ppm)
SnO
OSn
O
O
O
O
O
O
O
O
O
O
O
OO
Om
m
+ (2m+1) L-LAn n
initiator 4
t = 15 min
t = 1345 min
a
b
b
a
a’
a’
LLA
PLLA
0
20
40
60
80
100
0 500 1000 1500 2000 2500 3000
time [min]
conv
ersi
on [%
]
CL, initiator 4
LLA, initiator 4
LLA, initiator 5
-
Functionalized polyesters
37
appeared at 4.72 ppm and the peak at 4.35 ppm originates from
the –O–CH– protons directly attached to the tin atom.
Polymerization using 5 reveled the same results, peaks from the
initiator were seen in NMR spectra of the polymers. The results
confirmed that the alkoxide groups in all the initiators were
completely reacted under these reaction conditions, which is
consistent with earlier investigations with a similar initiator.61,
150
The progress of these reactions was also followed by SEC. In
every polymerization, the MWD of PLLA and PCL was narrow,
indicating fast initiation with respect to propagation. It also
indicates fast propagation compared to chain transfer or other
adverse termination reactions. The molecular weight increased
linearly with the conversion, which supports these statements. A
typical polymerization course using 4 or 5 as initiator is
presented in Figure 5.6.
Figure 5.6 Results from LLA polymerization using 4 as initiator.
Mn was determined by SEC using narrow polystyrene standards.
[M]/[I]=50, [M]0=0.5M, T=60°C
Figure 5.7 shows a semilogarithmic diagram of the data obtained
during the polymerizations. The straight lines confirm the previous
results, that the reactions are proceeding in a controlled way.
0
4000
8000
12000
16000
0 20 40 60 80 100
conversion [%]
Mn
1,02
1,05
1,08
1,11
1,14
1,17
MW
D
-
Functionalized polyesters
38
A)
B)
Figure 5.7 Semilogarithmic plot of – ln([M]/[M]0) versus
reaction time for polymerizations of LLA and ε-CL using the two
functionalized tin initiators, 4 and 5. [M]/[I]=50, [M]0=0.5M,
T=60°C
A number of polymerizations were performed using 4 and 5 as
initiators and LLA or ε-CL as monomer. The target DP was selected
to be between 20 and 500, which was controlled by adjusting the
molar ratio of monomer to initiator. Table 5.1 lists the
results.
Table 5.1 Results from the polymerizations of LLA and ε-CL using
the functionalized initiators. A) PLLA, initiated by 4 B) PCL,
initiated by 4 C) PLLA, initiated by 5
[M]/[I]a Time (h)
DPb MWDc Conv.d (%)
Yield e (%)
20 6 21 1.06 98 72 50 16 53 1.07 98 86
100 28 130 1.05 90 83 250 71 270 1.10 78 68 500 140 550 1.09 83
76
[M]/[I]a Time (h)
DPb MWDc Conv.d (%)
Yielde (%)
50 2.5 55 1.22 92 83 100 5 108 1.23 95 93 250 15 260 1.24 98 93
500 30 530 1.23 99 90
0,0
1,0
2,0
3,0
4,0
5,0
0 500 1000 1500 2000
time [min]
-ln[M
]/[M
] 0
CL, initiator 4LLA, initiator 4LLA, initiator 5
-
Functionalized polyesters
39
C)
a) molar feed ratio calculated from the unimeric species b)
calculated from 1H-NMR on precipitated polymer c) determined by SEC
analysis calibrated with narrow polystyrene standards d) conversion
obtained from crude samples before precipitation e) amount of
polymer formed after precipitation
The find DP agreed well with the monomer-to-initiator ratio; and
the MWD was narrow. To determine DP by quantification of the 1H-NMR
spectrum, the signals in the polymer and the signal from the
incorporated double bond were used. This excellent control of molar
mass indicates that the initiator efficiency is high with little
loss of active sites.
5.3 Epoxidation of the incorporated doublebond
Non-linear polymers have many structural variables (composition,
backbone length, branch length, branch spacing, etc.) that give a
great potential for new properties.151 To make a graft polymer with
specific properties, the structure must be controlled. The
macromonomer method is one of the most useful ways to design and
obtain well-defined graft copolymers in a controlled way. The
previous sections described how a double bond was incorporated into
the polyester chain backbone. The double bond is a perfect grafting
point where the distance between interconnecting points can be
precisely regulated by the DP. In order to test the reactivity of
the unsaturated group, epoxidation has been carried out at room
temperature with m-chloroperoxybenzoic acid (mCPBA) in chloroform,
Scheme 5.1. Aliphatic polyesters containing epoxides are
interesting since they can be used as precursors of degradable
networks.152 mCPBA was used because it is commercially available,
stable in solution at a moderate temperature for prolonged periods
and reacts under mild conditions.
Scheme 5.1 Epoxidation of the double bond using mCPBA.
[M]/[I]a Time (h)
DPb MWDc Conv. d(%)
Yielde (%)
20 8 21 1.07 95 40 50 16 51 1.11 96 83
100 32 100 1.11 95 78 500 160 503 1.08 65 63
HO
OO O
OO
HO
O
O
On n
HO
OO
O
On O
OO
HO
O
n
O
O O H
OCl
(mCPBA)
-
Functionalized polyesters
40
H OO
OO
O
On
OO
O HO
n
H OO
OO
On O
OO
O HO
n
O
g
h
i
j
The epoxidation was monitored by the use of NMR spectroscopy.
Figure 5.8 shows 1H-NMR